Arntzen, Tonnessen, & Brouwer (2006) Abberant Verbal Behavior

Behavioral Interventions
Behav. Intervent. 21: 177–193 (2006)

Published online in Wiley InterScience
(www.interscience.wiley.com) DOI: 10.1002/bin.220
REDUCING ABERRANT VERBAL BEHAVIOR BY BUILDING

A REPERTOIRE OF RATIONAL VERBAL BEHAVIORy
Erik Arntzen1, Ingvil Ro Tønnessen2 and Grete Brouwer2

1
Akershus University College, Norway
2
Community-based services, Kristiansand, Norway
The purpose of the present study was to train different verbal classes in a 44-year-old woman to reduce
aberrant verbal behavior. She had been hospitalized for 22 years in different psychiatric institutions, and
has been diagnosed with schizophrenia, developmental disabilities, non-organic psychosis, and autism.
Assessment of her verbal behavior showed that a small amount was adequate verbal behavior, whereas
almost the entire repertoire of verbal behavior could be characterized as aberrant verbal behavior
(‘psychotic’ and repetitive verbal behavior), and this behavior was maintained by social reinforcement
such as smiles, various comments from other persons. The treatment program was based on results from
the assessment results and was focused on establishing relationships between a variety of verbal response
responses and proper controlling variables. She was trained to produce different verbal classes, mainly
mand, tact, intraverbal, textual, transcription, and dictation responses. The duration of aberrant verbal
behavior and the rate of ‘psychotic’ verbal behavior were gradually reduced as a function of an
increasing number of adequate verbal responses. In the study, we present data covering a very long
period; the last data point was acquired 25 months after the start of the program. Copyright # 2006 John
Wiley & Sons, Ltd.
INTRODUCTION
Verbal behavior is regarded as operant behavior and may be defined as behavior

reinforced through the mediation of other persons (Skinner, 1957). A taxonomy for
different kinds of verbal behavior is dependent upon function and not structure, thus,
each category emphasizes an important facet of the contingency responsible for that
classification. Some categories may call attention to the reinforcer, others may
emphasize the antecedent conditions, while still others may highlight the relation
between antecedent conditions and the form of the response, and so on. Skinner
described seven primary classes: mand, tact, intraverbal, echoic, textual, transcrip-
tion, and dictation. A mand is a verbal operant under the control of deprivation or an
*Correspondence to: E. Arntzen, Akershus University College, P.O. Box 423, 2001 Lillestrøm, Norway.
E-mail: [email protected]
y
Some results from the present study were presented as a poster at the ABA convention in Boston 2004.
Copyright # 2006 John Wiley & Sons, Ltd.

178 E. Arntzen et al.
aversive stimulus and a response that characterizes the reinforcer. An establishing

operation is necessary for the occurrence of this verbal operant. It is like a boy asking
‘Can I have an apple?’ and the consequence being he gets an apple. The term mand
has been derived from words like command or demand. A tact is a verbal response that
occurs in the presence of a non-verbal stimulus and is maintained by social
reinforcers. This non-verbal stimulus is an environmental event, a property or a
relation between events. It is like a child saying ‘That is an apple’ in the presence of an
apple and the consequence being, for example, praise: ‘Good talking.’ The word tact
has been derived from the word contact. Verbal behavior is behavior that is under the
control of verbal stimuli and could be classified as to whether or not there is a one-to-
one correspondence. Intraverbal behavior is behavior that is under the control of
verbal stimuli, but there is not an accurate accordance between the verbal stimulus
and the verbal response. It is like answering ‘apple’ when asked ‘What is Granny
Smith?’ The other types of verbal behavior under the control of verbal stimuli are the
formal classes presented in Table 1, a schematic presentation Catania (1998) has
presented earlier. Echoic behavior is like saying ‘apple’ in the presence of the vocal
stimulus ‘apple.’ Dictation is writing ‘apple’ in the presence of the vocal stimulus
‘apple.’ Textual is reading ‘apple’ in the presence of the written stimulus ‘apple.’
Transcription is writing ‘apple’ in the presence of the written word ‘apple.’ From
these examples, it should be clear that it is impossible to assess the form or topography
of a verbal response, for example, apple, without knowing anything about the
functional relations controlling the verbal response.
Verbal responses that concern or trouble listeners are variously labeled as
perseverative (i.e., repetitive or stereotyped words or phases), delusional (obviously
false statements), psychotic (words or phrases unrelated to ongoing environmental
events), hallucinatory (verbal responses to unobservable stimuli), and echolalic
(persistent repetitions of the words or phrases of others) (Ewing, Magee, & Ellis,
2002; Wilder, Masuda, O’Connor, & Baham, 2001). These behavior problems are
common in developmentally disabled and psychiatric populations, and such bizarre or
aberrant speech is sometimes considered to be symptomatic of an underlying mental
illness, such as schizophrenia. Thus, traditional content analyses have focused on
Table 1. The table shows four different types formal classes dependent upon the combination of vocal or
written as discriminative stimulus (Sd) or verbal response (RV).
Sd/RV Vocal Written
Vocal Echoic Dictation

Written Textual Transcription
Note: Some have argued that the controlling variables for transcription may be more complex (e.g., Skinner, 1957)
than what is presented in this table, in the sense that when copying a text people first make an overt or covert textual
response, and then a transcription response to their textual response.
Copyright # 2006 John Wiley & Sons, Ltd. Behav. Intervent. 21: 177–193 (2006)
DOI: 10.1002/bin
Reducing aberrant verbal behavior 179
word classes, for example, nouns, verbs, adjectives, etc. (see for example Rosenberg
& Tucker, 1979), while Skinner (1957) argued that verbal behavior could be analyzed
like any other operant behavior. Furthermore, researchers have experimentally
examined variables that maintain psychotic, bizarre, or delusional verbalization and
demonstrated the operant nature of these responses (e.g., Isaacs, Thomas, &
Goldiamond, 1960; Mace, Webb, Sharkey, Matson, & Rosen, 1988). But there also
the question of whether the responses are maintained by automatic reinforcement
(Lancaster et al., 2004).
Aberrant or bizarre talk is a behavioral problem not often focused on in the
functional literature, even though there have been some important contributions
(DeLeon, Arnold, Rodriguez-Catter, & Uy, 2003; Dixon, Benedict, & Larson, 2001;
Dunlap, Kern-Dunlap, Clarke, & Robbins, 1991; Durand & Crimmins, 1987; Ewing
et al., 2002; Lancaster et al., 2004; Mace & Lalli, 1991; Mace et al., 1988; Wilder
et al., 2001). We wished to contribute to the literature by using a descriptive analysis
to reveal functional relations.
Behavioral interventions have included (a) withholding attention in response to
bizarre statements (e.g., Ayllon & Michael, 1959), (b) differential reinforcement of
rational speech (e.g., Bartlett, Ora, Brown, & Butler, 1971; Dixon et al., 2001; Dixon
et al., 2004), (c) non-contingent attention (Lancaster et al., 2004), (d) punishment of
bizarre verbal responses—time-out (e.g., Reichle, Brubakken, & Tetreault, 1976),
and (e) some combinations of these treatment elements (e.g., Mace et al., 1988). We
wished to expand this list of procedures by showing that, through establishment of a
verbal repertoire with different verbal operants, it was possible to reduce aberrant
verbal behavior. Furthermore, the curriculum for the treatment program for training
different classes of verbal behavior was based on an assessment of verbal behavior
using The Assessment of Basic Language and Learning Skills (ABLLS, see
Partington & Sundberg (1998)).
Therefore, the purpose of the current study was twofold: (1) to assess the verbal
repertoire and provide a descriptive analysis of the participant’s verbal behavior, and
(2) to reduce psychotic or aberrant talk by training the participant in different classes
of verbal behavior and through extinction of aberrant verbal behavior.
METHOD
Participant
Ann, a 44-year-old woman diagnosed as schizophrenic, autistic, developmentally
disabled, and as periodically psychotic—F28 (non-organic psychosis) (ICD-10), took
part in the present study. Her intellectual level has been difficult to determine due to
DOI: 10.1002/bin
non-compliance1. Because of her extensive non-compliance behavior, the training

had to be arranged in a certain way by including some very easy tasks (described
later). She had spent 22 years in a psychiatric institution, and in 1998 she moved into
her own apartment in a residential center that provides services to six mentally
disabled adults. Further, she has been under post-prison surveillance since 1985
(arson conviction), and she has now been sentenced to preventive supervision. Thus,
Ann has her own staff and is under constant supervision. In addition to aberrant verbal
behavior, she has displayed violent threats, destroying her furniture and belongings,
physically attacking the staff, and so on. She was also an excessive smoker, which
suggested that access to tobacco could be a powerful reinforcer (described later).
Definitions and Recordings of Target Behavior

Aberrant behavior was used as a label for psychotic and repetitive verbal behavior.
Psychotic verbal behavior was defined as talk about whether she is going to die within
an hour or so and other mathematic calculations on this theme, loss of her security,
what type of liquid her body contains, whether this place is going to be empty in a few
seconds, talk about dead bodies, and so on. Repetitive behavior was defined as
repetition of sequences, more than once, of vocal verbal behavior within a 30-s
period. Quiet time or when she was silent was defined as time during which she did
not produce any vocal verbal behavior. We found quiet time to be important, because
often instead of listening when people talked to her she produced aberrant behavior.
Thus, both producing more rational verbal behavior and being quiet seemed to be
significant.
The different classes of rational verbal behavior were defined as follows: If she said
‘Can I have some coffee’ and the consequence was that she was offered coffee, we
defined it as a mand. If she said, ‘It is a red cup’ in the presence of a red cup and it
produced a consequence like ‘Yes, it is a red cup,’ we defined it as a tact. If she said ‘It
is hot outside’ in the presence of the verbal stimulus ‘It is hot outside’ and it produced
a consequence like ‘Yes, that’s correct,’ we defined it as an echoic. If she said
‘Helsinki’ in the presence of the verbal stimulus ‘Tell me the capital city of Finland’
and it produced a consequence like ‘Yes, Helsinki is the capital city of Finland,’ we
called it an intraverbal. If she wrote ‘There are four seasons of the year, winter, spring,
summer, and autumn’ in the presence of the written text ‘There are four seasons of the
year, winter, spring, summer, and autumn’ and it produced a consequence like
‘Correct, good writing,’ we defined it as a transcription. If she wrote ‘I’m living in a
city in the southern part of Norway’ in the presence of the vocal stimulus ‘I’m living
1
Before the training program started she had been referred to as non-testable, but at the end of the current study, she
was successfully tested using WAIS and had a score of 54.
DOI: 10.1002/bin
in a city in the southern part of Norway’ and it produce a consequence like ‘Correct,
good writing,’ we defined it as dictation. If she read, ‘The parliament of Norway has
169 representatives’ in the presence of the text ‘The parliament of Norway has
169 representatives’ and it produced a consequence like ‘Correct, good reading,’ we
defined it as a textual.
Four types of recordings were made: (1) We recorded duration of quiet time,
rational and aberrant (psychotic and repetitive) verbal behavior on a minimum of six
occasions a week, at least 20 min per occasion. The therapists used two stop watches
to record the verbal responses categorized as aberrant and rational, and quiet time was
scored as the difference between 20 min and the sum of the duration of aberrant and
rational responses. This recording contained both treatment sessions and extra-
treatment sessions. Quiet time was recorded to ensure that Ann did not emit any
noises, for example, squealing. We did not wish to reduce quiet time per se, but rather
replace aberrant behavior with rational behavior. We found that recording of duration
was appropriate, because the number of responses could increase, while the duration
remained the same or even decreased. (2) Because the recording was confounded by
both treatment and extra-treatment settings, we carried out a pre- and post-test similar
to (1), except that it was only done in extra-treatment settings. (3) In addition, we also
recorded frequency of psychotic verbal behavior by video recording at intervals of
20 min randomly selected throughout the day. We recorded this monthly. This
recording was made to study changes specific to psychotic verbal behavior, as this
behavior has been used as one of the criteria for diagnosing her as schizophrenic and
psychotic, and also because it was interesting to study whether such verbal behavior
could be influenced by environmental consequences. (4) Also, during the training
sessions, all verbal responses were recorded: a correct response was recorded as a plus
and an incorrect response as a minus. We recorded every training session. The
recordings were made throughout the day.
Baseline Conditions
Two staff members interacted with the participant, and the interaction conditions
were the same as they had always been. No programmed reinforcers were presented
for verbal behavior. The recordings of duration and rate of rational, aberrant and quite
time were made under the same conditions as later during the training program.
Assessment and Descriptive Analysis

We used the ABLLS to assess Ann’s verbal repertoire, and the results were used to
determine the curriculum for the training program.
Furthermore, a descriptive analysis was conducted to assess whether her aberrant
(repetitive or psychotic) verbal behavior was under the control of automatic
DOI: 10.1002/bin
reinforcement or being maintained by social reinforcers. We used direct observation

and a tape recorder to assess the frequency of different verbal responses under the
different conditions. Observation sessions were 30–45 min in duration and were
conducted at random times between the hours of 10:00 a.m. and 6:00 p.m. We
recorded the rate of verbal behavior during situations in which there was interaction,
no interaction but not alone, alone, constant attention, demand, attention, and no
programmed reinforcers.
Procedure
The treatment program was based on assessment results and focused on
establishing relationships between a variety of verbal responses and proper
controlling variables. The training sessions were held in Ann’s apartment. Ann
and the trainer sat opposite each other, with a table between them. Each training
session was designed to include different classes of behavior, for example, imitation,
instruction, tactile stimuli, intraverbals, and training of preposition skills. The
program was based on results from the ABLLS. The different operants were trained in
random order. In the beginning, each training session was relatively simple and
involved the same topic (e.g., fruit, furniture, clothes, and so on), except for imitation,
instruction and tactile stimuli, and gradually became more difficult over time. The
training involved presentation of both objects and photos. For example, one session
started with an imitation trial, followed by an echoic, mand compliance, tact,
transcription, dictation, intraverbal, textual, and ended with a mand trial. A session
similar to one at the end of the training program with different training trials is shown
in Table 2. Each training session lasted approximately 10 min. Ann could refuse to
participate in training, but a mean of 8 training sessions were completed each day,
Monday through Sunday. Responses such as echoics and imitation were chosen to
ensure that she at least had some correct responses and complied with participation in
the training sessions.
Correct responses were followed by verbal praise, according to a FR1 schedule
of reinforcement. Verbal praise has previously increased verbal responses in her
repertoire. The verbal praise was specified to ensure that aberrant behavior was not
reinforced, for example, ‘you are good at writing the word grape.’ After completing a
training session, Ann was given a token as reinforcement. When five tokens had been
collected, they could be exchanged for various items (e.g., money, cosmetics, and
snacks), which we had observed increased Ann’s compliance behavior in other
settings.
Tact responses were defined as incorrect if she, for example, said that fork was
under the table, when it was on the table. Intraverbal responses were defined as
incorrect if she answered ‘spring’ when asked which season Christmas falls in.
DOI: 10.1002/bin
Table 2. The table shows an example of classes of responses and instructions and materials within a
session at end of the treatment program.
Class Instructions and materials
Transcription ‘We eat crabs in the summer,’ ‘I have to remember to use suntan
lotion in the summer.’
Tact We used different magazines and turned to a random page
and asked her to describe what she saw.
Intraverbal We asked different questions regarding the same picture, e.g., ‘Which
class is a knife, which shape has an apple,’ ‘which season is
Christmas Time?’
Textual She was asked to read aloud four lines from the newspaper.
Intraverbal We asked her questions about the text she has read.
Tacting of gustatory stimuli She was going to name different tastes as sour, bitter, sweet, and salt.
Matching Different types of objects, pictures, letters, words, etc. were used.
Mand compliance ‘Go into the kitchen and bring with you something within the
class cutlery.’
Tacting of olfactory stimuli Different types of stimuli as instant coffee, chocolate, green soap,
popcorn, toothpaste, peanuts, honey etc. were presented when she had
her eyes closed and she was going tell the different smells.
Synonym ‘Give me synonym for running, having fun, box, etc.’
Tacting of tactile stimuli She was going to name different stimuli as: rough, smooth, solid,
soft, sticky, hairy, sharp, and wet.
Dictation She was going to write, ‘It is fun with summer time.’ If we were to
record it as correct she had to start with a capital and end with
punctuation.
Delayed labeling Five objects/pictures are presented on table in front of her. The
objects/pictures were from the tasks above. She was asked to name the
objects/pictures twice and to shut her eyes. Two of the objects/
pictures were then removed. Her task was to name the two objects/
pictures which were removed.
Prepositions She was asked to name objections in all different positions.
Intraverbal—remembering The trainer read a text and she asked questions from the text.
Textual responses were defined as incorrect if she read anything other than what was
presented in the text. Dictation responses were defined as incorrect if she, for
example, wrote words other than those dictated. Transcription responses were defined
as incorrect if there was not a complete accordance between what was presented in the
text and what she wrote. Thus, if she did not respond correctly, different types of
prompts, for example, pointing prompt, echoic prompt, verbal directions and
modeling were used. All types of prompted responses were recorded as incorrect
responses. In addition to the training program, we also trained under more natural
conditions, and she was asked a minimum of 25 questions during each 20-min period.
DOI: 10.1002/bin
There was a minimum of five such periods each day. These questions included
training components mention above, like tacting, textual behavior, intraverbal be-
havior, and so on. Every correct response was followed by verbal praise in descriptive
form as above, for example, FR1. All aberrant verbal behavior was extinguished
during training sessions and under natural conditions. Access to tobacco was
introduced as a consequence for compliance during training (after recording 32). It
was arranged in such way that she did not have less access to tobacco than before, but
the same amount was given as a consequence of training.
Extinction
An extinction procedure was applied in combination with the training of verbal
behavior. Attention or verbal comments were withheld when psychotic and repetitive
verbal behavior occurred, while social reinforcers were programmed when she
produced rational verbal responses.
Reliability
Three types of reliability were calculated: (1) Familiarity with verbal operants.
Tests for reliability were conducted to ensure that the trainers discriminated between
the different verbal operants. This was a paper-and-pencil test with 36 different text
examples from the participant’s repertoire, and the trainers were to write down the
correct verbal classes. All trainers took part in these tests every other week. For
duration measurements, the margin of error was 3–4 s allowed in defining an
agreement (same as for (2)). The highest reliability score was 100%, the lowest score
90.3%, and the mean was 93.9%. (2) Another test, video clips, was administered to
ensure that the trainers recorded the same duration of quiet time, rational and aberrant
verbal behavior. This test was conducted every other week. The highest reliability
score was 95.7%, the lowest 91.2%, and the mean was 93.5%. (3) During training
sessions, interobserver agreement tests were administered during 58.4% of all
training sessions. This was done to ensure that the trainers measured the same correct
and incorrect response. Two persons (trainer and another staff) assessed correct and
incorrect responses during the training sessions. The highest reliability score was
100%, the lowest was 92%, and the mean 93.5%. For tests (1), (2), and (3), number of
agreements was divided by the total number of occurrences and multiplied by 100.
Furthermore, all trainers were trained in advance on the operational definitions.
Design
The data are presented in an AB design, and this is therefore a type of case study.
Threats against internal validity such as alternative interpretations of results for
DOI: 10.1002/bin
uncontrolled case studies have been discussed (e.g., Kazdin, 1982; Shadish, Cook, &
Campbell, 2002). There are limitations in drawing valid conclusions from case
studies such as the current one, but case studies could be improved in clinical practice
where methodological alternatives are unavailable (Kazdin, 1981). Improvement of
case studies by including specific features is discussed later. A reversal was
implemented because the number of training sessions decreased markedly during
a period of 4 weeks, 0.9 per day due to summer vacation. There were no
recordings during this period of low exposure to treatment, but there were recordings
following this period and immediately before the number of training sessions was
increased.
RESULTS
The results showed that she had good skills in some areas, for example, simple
textual and transcription in addition to comprehensive echoic behavior, but her
score was particularly low on different tact relations, for example prepositions,
and intraverbals. The analysis results showed that her aberrant vocal behavior
was maintained by social reinforcers, and mainly by attention. We found that the
rate of psychotic verbal behavior was reduced when the therapists left her alone
and that there were no psychotic verbal responses when she was alone in
the apartment. In contrast, when the therapists returned, the rate of psychotic verbal
behavior increased. Further, the results from the descriptive analyses showed that the
amount of rational verbal behavior compared to the total verbal repertoire was very
limited.
During baseline, the duration of quiet time per minute varied between 28.8 s and
44.4 s, with a mean duration of 37.8 s (see Figure 1). For aberrant (psychotic and
repetitive) verbal behavior, the duration per minute varied between 13.2 s and 28.8 s,
with a mean duration of 20.7 s. For rational verbal behavior, the duration per minute
varied between 1.2 s and 1.8 s, with a mean duration of 1.5 s. After implementation of
the intervention, aberrant verbal behavior decreased initially and then increased
slowly until the 20th recording. One can see the opposite trend in the duration of quiet
time. Also the duration of rational verbal behavior is increasing. From the 20th
recording, we see a slightly decreasing trend for aberrant verbal behavior. As marked
in Figure 1 by a larger legend on the thirtieth recording, there is an increase in the
duration of aberrant verbal behavior and a decrease in the duration of rational verbal
behavior following the 4-week period with very low exposure to treatment. Following
ordinary exposure to the treatment program, we can see in Figure 1 that the duration
of aberrant verbal behavior decreases, while the duration of rational verbal behavior
increases. We can also see for the first time for recording 33 that the duration of
DOI: 10.1002/bin
Quiet time, aberrant VB (psychotic and repetitive), and rational VB
60
Baseline
50
Quiet time
Mean Seconds Per Minute
40
30
aberrant vb
(psychotic and repetitive)
20
10
rational vb
0
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47
Recordings
Figure 1. The figure shows the duration of quiet time, aberrant (psychotic or repetitive) and rational
verbal behavior per minute. Each data point represents a mean of six recordings as described in the
method section. Until recording 30, every data point represents 1 week, and then until recording 39 every
data point represents every other week. The last 10 data points represent monthly recordings. The
numbers for the recordings do not correspond to the numbers in Figure 2. The first recording after the 4-
week period with very low exposure to the treatment program is marked with an increased font for the
legend, the 30th recording.
rational verbal behavior is higher than the duration of aberrant verbal behavior. The
duration of quiet time is about the same at the beginning and the end of the treatment
program. On the last data point, the duration of rational verbal behavior is 18 s
compared to a mean of 1.5 s during baseline, and likewise the rate of aberrant verbal
behavior is 6 s compared to a mean of 20.7 s during baseline.
As can be seen in Figure 2, the mean rate of psychotic verbal behavior is 0.8 during
baseline, and decreasing during intervention, with a mean rate of 0.2. The seventeenth
data point is the first data point following the 4-week period, and an increase in
psychotic verbal behavior is observed. For the subsequent recording, there is a
decrease in the rate of psychotic verbal behavior and then an increase in rate for three
recordings, followed by a decrease in rate. The last recording shows a rate of 0.09 per
minute compared to the baseline rate of 0.8 per minute.
We have chosen to show results for changes in some tact responses (other results
are available upon request from the authors), because the pattern is representative of
DOI: 10.1002/bin
psychotic verbal behavior

1
0.9
rate (number of responses per min) 0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
1 3 5 7 9 11 13 15 17 19 21 23 25 27
Recordings
Figure 2. The figure shows the rate of psychotic verbal behavior for a period of 27 months. Each data
point represents the rate of responding per monthly recording.
the other responses trained. The number of correct tact responses in the upper panel of
Figure 3 (prepositions) increases from the first part to the second part of training (for
convenience, we have only divided the training into two parts), because the number of
incorrect responses decreased. There was also an increasing number of objects and
pictures, and requirement of more difficult relations between those. In the lower panel
showing correct and incorrect delayed labeling responses, as well as progression in
the training program (i.e., both retention interval and number of objects and pictures),
there is an increase in number of correct responses and a decrease in number of
incorrect responses.
Recordings from the pre-test in the extra-therapy settings showed that the duration
of rational verbal behavior was 0.96 s per minute and of aberrant verbal behavior
7.68 s per minute. In the post-test, the duration of rational verbal behavior was 10.32 s
per minute and of aberrant verbal behavior 0.99 s per minute.
DISCUSSION
The purpose of the current study was to build a verbal repertoire based on the
ABLLS assessment as well as, through this process, to reduce aberrant behavior that,
according to an assessment of the descriptive analysis, was being maintained by social
DOI: 10.1002/bin
Increasing number of objects and

Tact responses pictures, and also requirement of
more difficult relations between
600 those.
500
400
Number of responses
correct
300
200 incorrect
100
0
1 2
Parts of training
correct Tact responses - Delayed labeling

500
Both retention
interval and
400 number of
objects and
pictures are
increased.
Number of responses
300
200
Incorrect
100
0
1 2
Parts of training
Figure 3. The upper panel of the figure shows number of correct and incorrect prepositions responses
during training. From part 1 to part 2 of training number of objects and pictures trained increased; there
was also an increase in requirement of more difficult relations between objects and pictures. The division
into two parts was made for the sake of convenience. The bottom panel of the figure shows number of
correct and incorrect delayed labeling responses during training. From part 1 to part 2 of training, both
retention interval and number of objects and pictures increased.
DOI: 10.1002/bin
reinforcement. The results showed that, as a function of an increasing repertoire of

verbal behavior, the duration of aberrant verbal behavior and the rate of psychotic
behavior were reduced.
After implementation of the intervention, the increase in rational verbal behavior
does not greatly affect the amount of aberrant behavior, although the mean number of
seconds per minute of aberrant behavior is lower than during baseline. One could
argue that this is the result of the extinction procedures rather than the training, at least
until the summer break recording 29. There is not much evidence of rational speech
replacing aberrant speech until recording 33. Therefore, it could be argued that
rational and aberrant speech are in the same response class, such that prompting and
reinforcement of rational speech tended to indirectly increase aberrant speech
(recordings 1–32), until the duration of rational speech was high enough to begin
replacing aberrant speech (recording 33). Another reasonable explanation is that we
introduced access to tobacco as a reinforcer after recording 32, which means that
she had to produce rational vocal behavior to have access to tobacco, and we can see
from Figure 1 that recording 33 is the first recording where number of seconds per
minute with rational speech was higher than aberrant speech.
The present results replicate previous findings indicating that differential
reinforcement may be effective in reducing bizarre or aberrant vocalizations and
increasing rational verbal behavior (Ayllon & Michael, 1959; Bartlett et al., 1971;
Dixon et al., 2001; Liberman, Teigen, Patterson, & Baker, 1973; Wilder et al., 2001).
Also, our results are in accord with arguments that abnormal and normal behaviors
can be viewed as reciprocally related, and the notion that psychotics can be seen as
presenting considerable abnormal behavior, or little normal behavior (Isaacs et al.,
1960). Furthermore, it is interesting to note that behavior classified as psychotic vocal
verbal behavior could be changed. Historically, such behavior, if accompanied by
other diagnostic criteria, has often resulted in labeling and treating the participant as
schizophrenic.
We do not have any reversal design for various reasons, mainly because in this
treatment program we wished to establish a verbal behavioral repertoire, and
obviously repertoires such as this are difficult, if not impossible, to reverse. We are
aware of the threats to internal validity in the current case study, but due to the
difficulties both with reversal and with multiple baseline design in some programs,
such as in building a repertoire of verbal behavior, we have used an AB design.
Naturally, such a design can never replace experimentation. On the other hand, the
first recording after a 4-week period of little exposure to the treatment shows that the
rate of aberrant verbal behavior increased and rational verbal behavior decreased, and
that as soon as treatment exposure returned to its initial level, the rate of aberrant
verbal behavior decreased while the rate of rational verbal behavior increased once
more. Some types of behavior are not able to satisfy the requirement that the
DOI: 10.1002/bin
experimenter can turn them on/off (Baer, Wolf, & Risley, 1968), such behaviors could
be verbal behavior, swimming etc. Accordingly, it is not unexpected to find that the
rate does not return to baseline level when the treatment program is terminated, or as
here reduced to a minimum, as the target responses in the present treatment program
are verbal behavior. Moreover, in order to increase our ability to draw valid inferences
from case studies (Kazdin, 1981), we have strengthen the current study first by
collecting objective and not anecdotal data, as anecdotal data often constitute the
weakness of case studies. Second, we use continuous assessment rather than only
post-intervention and/or pre-intervention assessments, which may strengthen the
internal validity. Third, regarding past and future projection, in the current study the
participant has shown aberrant verbal behavior as part of her verbal repertoire for at
least 5 years prior to implementation of the treatment, and there is nothing to indicate
that this would have changed if treatment had not been implemented. Given that
the different types of verbal behavior have been stable for an extended period and that
the changes coincide with the treatment, there is support for the notion that the
intervention may have led to the changes observed in the current study.
In the current study, the focus has been on building a verbal repertoire to reduce
aberrant verbal behavior, and the results showed that this has been a gradual process.
It seems quite clear that one would not expect to see abrupt changes in the verbal
behavior repertoire. Variables that might affect this gradual process are the
participant’s age, learning ability and history of aberrant verbal behavior. Most
studies focused on establishing a verbal repertoire have been conducted with children
(e.g., Charlop, Schreibman, & Thibodeau, 1985; Greer, Yaun, & Gautreaux, 2005;
Partington & Bailey, 1993; Partington, Sundberg, Newhouse, & Spengler, 1994;
Yamamoto & Mochizuki, 1988). Here, treatment programs based on behavior
analysis have shown to be very effective (for a review of establishing verbal
repertoires see O’Neill, 1990). Concerning the age of participants, there are some
exceptions, one of them a study by Green, Linsk, and Pinkston (1986) on elderly
participants in which positive verbal responses were reinforced and undesired verbal
responses were ignored. Also Dixon et al. (2004) have implemented differential
reinforcement of alternative verbal behavior to reduce instances of inappropriate
verbal behavior in four adults with mental retardation. However, none of these studies
was carried out to establish a verbal repertoire per se.
During the present treatment program, the proportion of quiet time was fairly
constant at 40–50%, which is normal if we consider an ordinary conversation between
two people. On the other hand, we have observed that the quality of quiet time
changed during the treatment program, becoming more characteristic of listening
behavior. We intend to focus more on this aspect of her repertoire in the future.
Further, future efforts will include introducing training on autoclitic behavior, which
is verbal behavior that is under the control of the speaker’s own primary verbal
DOI: 10.1002/bin
behavior (mands, tacts, verbal behavior under control of verbal stimuli), which
influences the effects of that verbal behavior on the listener.
Social validity is an important aspect to consider whenever one is evaluating
treatment effects. The results of both duration and rate of aberrant behavior showed
that there was a decrease, and the participant was gradually more engaged in rational
verbal behavior when both treatment and extra-treatment sessions were taken
together. More specifically, the results from the pre- and post-test in the extra-therapy
settings showed an increase in duration of rational verbal behavior and a decrease in
duration of aberrant verbal behavior, which indicates that she is engaged in more
rational verbal behavior in natural settings. Furthermore, we have some information
showing that the participant produced appropriate vocalizations spontaneously and
that she used trained utterances in a socially meaningful way. After the treatment
program, she has more knowledge about intraverbals and the length of her
conversations has increased. For example, earlier she usually answered with one or
two words, and she did not continue the conversation. She is also a more active partner
in the conversation, with more turn-takings. Much of the aberrant verbal behavior is
now repetitions of more rational themes, while earlier it was more likely to contain
psychotic themes. She has also improved her writing skills in transcription and
dictation, as well as in a more functional way, for example writing postcards,
messages with correct grammar and spelling, and using periods. She also showed
more functional use of prepositions, synonyms and so on. Also, some other anecdotal
reports are rather positive, for example, people who have known Ann for quite a long
time ago have mentioned that her verbal repertoire has changed rather dramatically
and that she also behaves more appropriately when she meets people, while shopping,
etc. Also, to illustrate that she has changed her repertoire, before implementation of
the treatment program, it was not possible to accomplish a WAIS test, but at the end of
the current study she was successfully tested using WAIS.
In conclusion, the current study showed that through extinction of psychotic speech
and training to produce different verbal classes, mainly mand, tact, intraverbal,
textual, and dictation responses or differential reinforcement of alternative behavior
(rational talk), the duration of aberrant behavior and the rate of psychotic talk
decreased, while the duration of rational talk increased. There was a remarkable
change in the amount of aberrant versus rational verbal behavior when access to
tobacco was used as a reinforcer for producing rational talk during training. The
results from the current study show that psychotic behavior may be learned behavior.
ACKNOWLEDGEMENTS
Thanks to Rebekah Gauslaa Harring, Aina Engesland, Venke McCready, Liv-

Solveig Tønnessen, and Charlotte Loland Gaustad for their assistance in the treatment
DOI: 10.1002/bin
program. We are grateful to two anonymous reviewers for valuable comments and
suggestions on an earlier version of the manuscript.
REFERENCES
Ayllon, T., & Michael, J. (1959). The psychiatric nurse as behavioral engineer. Journal of the
Experimental Analysis of Behavior, 2, 323–334.
Baer, D. M., Wolf, M. M., & Risley, T. R. (1968). Some current dimensions of applied behavior analysis.
Journal of Applied Behavior Analysis, 1, 91–97.
Bartlett, D., Ora, J. P., Brown, E., & Butler, J. (1971). The effects of reinforcement on psychotic speech in
a case of early infantile autism, age 12. Journal of Behavior Therapy and Experimental Psychiatry, 2,
145–149.
Catania, A. C. (1998). Learning ( 4th ed.). Upper Saddle River, New Jersey: Prentice-Hall.
Charlop, M. H., Schreibman, L., & Thibodeau, M. G. (1985). Increasing spontaneous verbal responding
in autistic children using a time delay procedure. Journal of Applied Behavior Analysis, 18, 155–166.
DeLeon, I. G., Arnold, K. L., Rodriguez-Catter, V., & Uy, M. L. (2003). Covariation between bizarre and
nonbizarre speech as a function of the content of verbal attention. Journal of Applied Behavior
Analysis, 36, 101–104.
Dixon, M. R., Benedict, H., & Larson, T. (2001). Functional analysis and the treatment of inappropriate
verbal behavior. Journal of Applied Behavior Analysis, 34, 361–364.
Dixon, M. R., Guercio, J., Falcomata, T. S., Horner, M. J., Root, S., Newell, C., & Zlomke, K. (2004).
Exploring the utility of functional analysis methodology to assess and treat problematic verbal
behavior in persons with acquired brain injury. Behavioral Interventions, 19, 91–102.
Dunlap, G., Kern-Dunlap, L., Clarke, S., & Robbins, F. (1991). Functional assessment, curricular
revisions, and severe behavior problems. Journal of Applied Behavior Analysis, 24, 387–397.
Durand, M., & Crimmins, D. B. (1987). Assessment and treatment of psychotic speech in an autistic
child. Journal of Autism and Developmental Disorders, 17, 17–28.
Ewing, C. B., Magee, S. K., & Ellis, J. (2002). The functional analysis of problematic verbal behavior.
The Analysis of Verbal Behavior, 18, 51–60.
Green, G. R., Linsk, N. L., & Pinkston, E. M. (1986). Modification of verbal behavior of the mentally
impaired elderly by their spouses. Journal of Applied Behavior Analysis, 19, 329–336.
Greer, R. D., Yaun, L., & Gautreaux, G. (2005). Novel dictation and intraverbal responses as a function
of a multiple exemplar instructional history. The Analysis of Verbal Behavior, 21, 99–116.
Isaacs, W., Thomas, J., & Goldiamond, I. (1960). Application of operant conditioning to reinstate verbal
behavior in psychotics. Journal of Speech and Hearing Research, 25, 8–12.
Kazdin, A. E. (1981). Drawing valid inferences from case studies. Journal of Consulting and Clinical
Psychology, 49, 183–192.
Kazdin, A. E. (1982). Singe-case research designs. New York: Oxford University Press.
Lancaster, B. M., LeBlanc, L. A., Carr, J. E., Brenske, S., Peet, M. M., & Culver, S. J. (2004). Functional
analysis and treatment of the bizarre speech of dually diagnosed adults. Journal of Applied Behavior
Analysis, 37, 395–399.
Liberman, R. P., Teigen, J., Patterson, R., & Baker, V. (1973). Reducing delusional speech in chronic,
paranoid schizophrenics. Journal of Applied Behavior Analysis, 6, 57–64.
Mace, F. C., & Lalli, J. S. (1991). Linking descriptive and experimental analyses in the treatment of
bizarre speech. Journal of Applied Behavior Analysis, 24, 553–562.
DOI: 10.1002/bin
Mace, F. C., Webb, M. E., Sharkey, R. W., Matson, D. M., & Rosen, H. S. (1988). Functional analysis and
treatment of bizarre speech. Journal of Behavior Therapy and Experimental Psychiatry, 19, 289–296.
O’Neill, R. E. (1990). Establishing verbal repertoires: Toward the application of general case analysis
and programming. The Analysis of Verbal Behavior, 8, 113–126.
Partington, J. W., & Bailey, J. S. (1993). Teaching intraverbal behavior to preschool children. The
Analysis of Verbal Behavior, 11, 9–18.
Partington, J. W., & Sundberg, M. L. (1998). The assessment of basic language and learning skills (The
ABLLS): An assessment, curriculum guide, and skills tracking system for children with autism or other
developmental disabilities. Pleasant Hill, CA: Behavioral Analysts, Inc.
Partington, J. W., Sundberg, M. L., Newhouse, L., & Spengler, S. M. (1994). Overcoming an autistic
child’s failure to acquire a tact repertoire. Journal of Applied Behavior Analysis, 27, 733–734.
Reichle, J., Brubakken, D., & Tetreault, G. (1976). Eliminating perservarative speech by positive
reinforcement and time-out in a psychotic child. Journal of Behavior Therapy and Experimental
Psychiatry, 7, 179–183.
Rosenberg, S. D., & Tucker, G. J. (1979). Verbal behavior and schizophrenia. Archives of General
Psychiatry, 36, 1331–1337.
Shadish, W. R., Cook, T. D., & Campbell, D. T. (2002). Experimental and quasi-experimental designs for
generalized causal inference. Boston: Houghton Mifflin Company.
Skinner, B. F. (1957). Verbal behavior. Englewood Cliffs, NJ: Prentice-Hall, Inc.
Wilder, D. A., Masuda, A., O’Connor, C., & Baham, M. (2001). Brief functional analysis and treatment
of bizarre vocalizations in an adult with schizophrenia. Journal of Applied Behavior Analysis, 34,
65–68.
Yamamoto, J., & Mochizuki, A. (1988). Acquisition and functional-analysis of manding with autistic
students. Journal of Applied Behavior Analysis, 21, 57–64.
DOI: 10.1002/bin

Arntzen, Tonnessen, & Brouwer (2006) Abberant Verbal Behavior

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Arntzen, Tonnessen, & Brouwer (2006) Abberant Verbal Behavior

Uploaded by

Copyright:

Available Formats

Behavioral Interventions

Behav. Intervent. 21: 177–193 (2006)

REDUCING ABERRANT VERBAL BEHAVIOR BY BUILDING

Erik Arntzen1, Ingvil Ro Tønnessen2 and Grete Brouwer2

Verbal behavior is regarded as operant behavior and may be deﬁned as behavior

Copyright # 2006 John Wiley & Sons, Ltd.

aversive stimulus and a response that characterizes the reinforcer. An establishing

Vocal Echoic Dictation

non-compliance1. Because of her extensive non-compliance behavior, the training

Deﬁnitions and Recordings of Target Behavior

Assessment and Descriptive Analysis

reinforcement or being maintained by social reinforcers. We used direct observation

Quiet time, aberrant VB (psychotic and repetitive), and rational VB

psychotic verbal behavior

rate (number of responses per min) 0.8

Increasing number of objects and

correct Tact responses - Delayed labeling

reinforcement. The results showed that, as a function of an increasing repertoire of

Thanks to Rebekah Gauslaa Harring, Aina Engesland, Venke McCready, Liv-

You might also like